LRRK2 gene mutations are a common cause of Parkinsons disease. The protein product of the gene has both kinase and GTPase activities. Because there are mutations in both kinase and GTPase domains, we consider that both activities are probably important for pathogenesis of Parkinsons disease. As such, we are trying to understand each activity in turn and how they interact. We have shown, in collaboration with Quyen Hoang at Indiana University, that mutations in the GTPase (ROC) domain prolong the time in which LRRK2 is in the active state. Using biochemical approaches, we have now shown that GTP hydrolysis to GDP drives dissociation of dimers, which we presume are inactive, into monomers, presumably active. Mutations that lower GTPase activity favor the monomeric state. This is reflected in accumulation of LRRK2 protein at the trans-Golgi network in cells. We have used similar assays to formally compare properties of mouse and human LRRK2. Perhaps surprisingly, the two homologues have quite different properties, with human protein being less stable and more active than the mouse counterpart. Part of our motivation here was to try and understand the effects of a risk factor variant that is relatively common in some populations, G2385R, but is not conserved between mouse and human LRRK2. We found that G2385R in human LRRK2 causes the protein to become less stable but that a similar substitution in mice, E2385R, had no effect that we could measure. These results may indicate why it has been challenging to model LRRK2 mutational effects in mice.